WindowSHAP: An efficient framework for explaining time-series classifiers based on Shapley values
WindowSHAP: An efficient framework for explaining time-series classifiers based on Shapley values
复制标题
WindowSHAP：基于 Shapley 值解释时间序列分类器的有效框架
DOI：
10.1016/j.jbi.2023.104438
复制
发表时间：
2023
影响因子：
4.5
通讯作者：
Subbian, Vignesh
中科院分区：
文献类型：
--
作者：
Nayebi, Amin;Tipirneni, Sindhu;Reddy, Chandan K.;Foreman, Brandon;Subbian, Vignesh
解开和理解黑盒机器学习算法（如深度学习模型）如何做出决策一直是研究人员和最终用户面临的一个持续挑战。解释时间序列预测模型对于具有高风险的临床应用是有用的，以理解预测模型的行为，例如，以确定不同的变量和时间点如何影响临床结果。然而，现有的方法来解释这些模型往往是独特的架构和数据的功能不具有随时间变化的组件。在本文中，我们介绍了WindowSHAP，一个模型无关的框架，用于解释时间序列分类器使用Shapley值。我们打算使用WindowSHAP来降低计算长时间序列数据Shapley值的计算复杂性，并提高解释的质量。WindowSHAP基于将序列划分为时间窗口。在此框架下，我们提出了三种不同的算法Stationary，Sliding和Dynamic WindowSHAP，每一种算法都使用扰动和序列分析指标对基线方法KernelSHAP和TimeSHAP进行了评估。我们将我们的框架应用于来自专业临床领域（创伤性脑损伤- TBI）和广泛临床领域（重症监护医学）的临床时间序列数据。实验结果表明，基于这两个定量指标，我们的框架是上级解释临床时间序列分类，同时也降低了计算的复杂性。我们表明，对于具有120个时间步（小时）的时间序列数据，与KernelSHAP相比，合并10个相邻时间点可以将WindowSHAP的CPU时间减少80%。我们还表明，我们的动态WindowSHAP算法更侧重于最重要的时间步骤，并提供更容易理解的解释。因此，WindowSHAP不仅加快了时间序列数据Shapley值的计算，而且还提供了更高质量的更易于理解的解释。
Unpacking and comprehending how black-box machine learning algorithms (such as deep learning models) make decisions has been a persistent challenge for researchers and end-users. Explaining time-series predictive models is useful for clinical applications with high stakes to understand the behavior of prediction models, e.g., to determine how different variables and time points influence the clinical outcome. However, existing approaches to explain such models are frequently unique to architectures and data where the features do not have a time-varying component. In this paper, we introduceWindowSHAP, a model-agnostic framework for explaining time-series classifiers using Shapley values. We intend forWindowSHAPto mitigate the computational complexity of calculating Shapley values for long time-series data as well as improve the quality of explanations.WindowSHAPis based on partitioning a sequence into time windows. Under this framework, we present three distinct algorithms ofStationary,SlidingandDynamic WindowSHAP, each evaluated against baseline approaches, KernelSHAP and TimeSHAP, using perturbation and sequence analyses metrics. We applied our framework to clinical time-series data from both a specialized clinical domain (Traumatic Brain Injury - TBI) as well as a broad clinical domain (critical care medicine). The experimental results demonstrate that, based on the two quantitative metrics, our framework is superior at explaining clinical time-series classifiers, while also reducing the complexity of computations. We show that for time-series data with 120 time steps (hours), merging 10 adjacent time points can reduce the CPU time ofWindowSHAPby 80 % compared to KernelSHAP. We also show that ourDynamic WindowSHAPalgorithm focuses more on the most important time steps and provides more understandable explanations. As a result,WindowSHAPnot only accelerates the calculation of Shapley values for time-series data, but also delivers more understandable explanations with higher quality.